Job Reallocation and Entrepreneurship: Decomposing Connecticut’s Recovery
“Thanks for keeping me alive.”
Searching for Sugar Man (2012)
Has Connecticut’s economy lost its dynamism? Such an inquiry is necessary to establish whether our lack of vitality may be a contributor to our slower than average recovery. Importantly, an assessment focuses the areas where more attention should be brought to bear if we are to reboot our entrepreneurial engine. Indeed, institutions of higher education like the University of New Haven are keenly positioned to aid this reboot by reducing labor market frictions, enhanced investment in human capital, improved matching of skill sets in labor markets, and actively scrutinizing entry and exit barriers and other regulatory impairments to a more dynamic state entrepreneurial environment.
The more quickly a faltering or sluggish economy can reallocate resources to the places where they are most productive the more quickly it rebounds. If new firms are able to quickly launch; or existing firms promptly expand; and unproductive firms quickly shrink or exit, then the economy as a whole will experience faster growth. This reallocation of resources is associated with increased productivity, and economic and wage growth.
There are two simple way to quantify an economy’s speed of reallocation. One is to examine the historical development of establishment births and deaths in the state economy; the second one is to examine the number of jobs created and the number of jobs destroyed by business establishments, what is known as job reallocation. Both measures offer considerable insight into the nature of an economy’s entrepreneurial dynamism.
We scrutinize both measures individually and compared to the United States at large, using data from the U.S. Census Bureau’s Business Dynamic Statistics. In this first installment, we focus on the state of Connecticut’s job reallocation performance individually and relative to the United States.
Job Reallocation: Connecticut
Job reallocation describes how quickly jobs are moving from shrinking or closing firms to opening or expanding firms in the economy. Specifically, Gross Job gains proceed from opening and expanding establishments. Gross Job Losses occur because establishments close or contract. Formally, Job Reallocation is equal to Job Gains plus Job Losses. Job Reallocation is a measure of the intensity of job churning.
Figure 1 displays the historical job reallocation performance for the state of Connecticut starting in the third Quarter of 1992 and ending in the third Quarter of 2016.
Figure 1 illustrates the dramatic decline in Connecticut’s labor market churn relative to the late 1990s. A decline in job reallocation capability could indicate less innovation or less labor market flexibility, both of which are likely to retard economic growth.
Figure 2 plots the same data as a ratio of job gains to job losses – what is known as the JCDR, short for Job Creation Destruction Ratio. A drop below the horizontal line at 1 implies that job losses exceed job gains.
The JCDR is larger than 1 in 21 out of the 29 reported quarters since the end of the great recession in the 2nd quarter of 2009. This indicates that Connecticut has experienced more job gains than job losses over 72 percent of the time. However, Connecticut is lagging when contrasted with the nation’s JCDR performance over the same period; e.g. Figure 5, below.
There are two ways to gain jobs. First, jobs can be created from establishments that expand and jobs can be created from newly formed establishments. The newly formed establishment speaks more directly to the entrepreneurial vigor of an economy. Conventional wisdom and economic understanding recognize that start-ups contribute much to job creation. Importantly, a growing, vigorous startup environment assures increased market competition and the faster replacement of lower-productivity business with new, more productive ones.
Figure 3 and Figure 4 provide a glimpse as to the Connecticut trend in startups. Figure 3 shows employment gained from startups and Figure 4, the percentage of jobs (relative to total jobs) gained from startups.
Again, we see a marked decline from the levels of the 1990s. The message conveyed by Figure 3 and 4 is that Connecticut relies increasingly less on establishment openings for job creation. New Jobs accounted for approximately 25 percent of total job gained during the early part of the 1990’s. Lately this proportion has declined to about 17 percent a 32 percent decrease over the period, or approximately 1.3 percent per year.
Job Reallocation: Connecticut Relative to the United States
Figure 5 displays both data series indexed to 100 in the 3rd Quarter of 1992. Following a peak in the late 1990’s Connecticut’s labor market churn has underperformed the United States. It is equally troubling to see a similar secular decline for the United States.
Figure 6 shows the relative share of jobs from Establishment Openings for both Connecticut and the United States. Whereas this ratio is consistently higher for Connecticut throughout the period examined, both similarly decline throughout. However, Connecticut’s decline is considerably more rapid and variable.
On Table 1 is data on Jobs Recovered since the great recession for the state of Connecticut – by sectors. The percentage in Column 2 reflects the amount of jobs that have been added between February 2010 and May 2017 as a percent of jobs lost between March 2008 and February 2010. In effect, the table is a tally of how well we are doing across the state economy.
JOB LOSS RECOVERY
TOTAL NONFARM EMPLOYMENT
Transportation & Public Utilities
Professional & Business Services
Education & Health Services
Leisure and Hospitality
Source: CT Department of Labor: here.
Note the contrast between our jobs performance in Goods producing versus jobs in the Service sector – they are highlighted in red. Most of the jobs recovered – by far - have been in the service sector. And within the service producing sectors, Leisure and Hospitality stands out in particular, with a massive 340.7% percent rally, – followed by Other Services with a whopping 208.3%. Education & Health Services is another standout; it’s performance is listed above as NA because it was the only sector that did not lose jobs in the great recession – in fact, it added jobs during the recession and has added even more since February 2010, the end of the recession in Connecticut.
Baumol’s cost disease warns of the difficulties the service sector has in deploying productivity enhancements. [for a refresher on Baumol’s Cost Disease, see here.] In sum, we are adding jobs in sectors that promise little productivity gains and possibly ensuring an anemic economic growth over the near horizon. My colleague John Rosen, analyst extraordinaire, poses an answer (rhetorically):
Is it possible that Nutmegers are more likely to take their new jobs at new firms? Is this a preference or an accommodation to the realities of the job market in CT? Or, (a more disquieting hypothesis) is it the case that new establishments in CT are more likely to be the kind that hire lots of labor -- pizza parlors or auto body shops, for instance, rather than "new economy," fast growth, tech-based, software-defined startups?
My colleague Murat Akgun – asks for solutions. Again – being rhetorical, for in fact, the work that both Akgun and Rosen do in their innovation and entrepreneurial cooker New Tech Haven – is indispensable. New Tech Haven is a private sector initiative with all the right ideas. Theirs is a sorely needed solution – but it needs to be multiplied a hundredfold for it to be effective.
Bloomberg 2016 US Innovation Index ranked Connecticut 7th. I’m not sure where is the disconnect between what they are seeing and what we just posted. But from where we sit, the metrics are telling and disappointing: the future of the Connecticut economy looks far from healthy. The seeming lack of dynamism reflects a slow pace of innovation, a decline in entrepreneurship, an inability to adapt to the changing circumstances. We appear to have retreated from the entrepreneurial spirit that propelled the State’s once great prosperity. Unless we can turn around the present trends they portend a future of even slower economic growth.
- We are grateful to John Rosen and Murat Akgun for thoughtful comments.
- A more parsed examination of the same material we cover here can be found in the CTDOL’s Business Employment Dynamics Quarterly Reports. The latest of the CTDOL Report’s is the 2016 Q:3 issue; this latest issue as well as older reports can be obtained here. An earlier and more thorough examination of Connecticut’s Labor Market Dynamics is David W. Kennedy and Nicholas Jolly, “Connecticut’s Labor Market Dynamics: Job Creation, Destruction, and Reallocation,” (CTDOL, Wethersfield, CT: August 2006); it can be obtained here. We retain their nomenclature and framework to provide analytical continuity.
A.E. Rodriguez & Brian Marks
University of New Haven
Department of Economics
July 9, 2017
 As of May 2017, Connecticut has recovered 79 percent of the jobs it had in March 2007; source CT Department of Labor, here. Connecticut’s change in employment rate – between 2007 and 2016 - is -1.9 percentage points (81.8% in 2007 and 79.9% in 2016), whereas the United States change is -2 percentage points (79.9% in 2007 and 77.9 in 2016; source: Pew Trusts, here.